en-multinerd-ner-upsampled
This model is a fine-tuned version of distilbert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0339
- Precision: 0.9370
- Recall: 0.9459
- F1: 0.9414
- Accuracy: 0.9919
- Per-precision: 0.9951
- Per-recall: 0.9967
- Per-f1: 0.9959
- Org-precision: 0.9359
- Org-recall: 0.9445
- Org-f1: 0.9402
- Loc-precision: 0.9708
- Loc-recall: 0.9752
- Loc-f1: 0.9730
- Dis-precision: 0.7131
- Dis-recall: 0.7404
- Dis-f1: 0.7265
- Anim-precision: 0.6809
- Anim-recall: 0.7187
- Anim-f1: 0.6993
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | Per-precision | Per-recall | Per-f1 | Org-precision | Org-recall | Org-f1 | Loc-precision | Loc-recall | Loc-f1 | Dis-precision | Dis-recall | Dis-f1 | Anim-precision | Anim-recall | Anim-f1 |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
0.0293 | 1.0 | 9223 | 0.0262 | 0.9302 | 0.9363 | 0.9333 | 0.9911 | 0.9928 | 0.9951 | 0.9939 | 0.9256 | 0.9357 | 0.9306 | 0.9673 | 0.9622 | 0.9647 | 0.6819 | 0.7286 | 0.7045 | 0.6710 | 0.6799 | 0.6754 |
0.0149 | 2.0 | 18446 | 0.0282 | 0.9353 | 0.9466 | 0.9409 | 0.9918 | 0.9948 | 0.9969 | 0.9959 | 0.9322 | 0.9448 | 0.9385 | 0.9687 | 0.9725 | 0.9706 | 0.7190 | 0.7457 | 0.7321 | 0.675 | 0.7381 | 0.7051 |
0.0084 | 3.0 | 27669 | 0.0339 | 0.9370 | 0.9459 | 0.9414 | 0.9919 | 0.9951 | 0.9967 | 0.9959 | 0.9359 | 0.9445 | 0.9402 | 0.9708 | 0.9752 | 0.9730 | 0.7131 | 0.7404 | 0.7265 | 0.6809 | 0.7187 | 0.6993 |
Framework versions
- Transformers 4.36.1
- Pytorch 2.1.0+cu121
- Datasets 2.15.0
- Tokenizers 0.15.0
- Downloads last month
- 3